\documentclass[10pt]{article}
\usepackage{fullpage}
\usepackage{setspace}
\usepackage{parskip}
\usepackage{titlesec}
\usepackage[section]{placeins}
\usepackage{xcolor}
\usepackage{breakcites}
\usepackage{lineno}
\usepackage{hyphenat}
\PassOptionsToPackage{hyphens}{url}
\usepackage[colorlinks = true,
linkcolor = blue,
urlcolor = blue,
citecolor = blue,
anchorcolor = blue]{hyperref}
\usepackage{etoolbox}
\makeatletter
\patchcmd\@combinedblfloats{\box\@outputbox}{\unvbox\@outputbox}{}{%
\errmessage{\noexpand\@combinedblfloats could not be patched}%
}%
\makeatother
\usepackage{natbib}
\renewenvironment{abstract}
{{\bfseries\noindent{\abstractname}\par\nobreak}\footnotesize}
{\bigskip}
\titlespacing{\section}{0pt}{*3}{*1}
\titlespacing{\subsection}{0pt}{*2}{*0.5}
\titlespacing{\subsubsection}{0pt}{*1.5}{0pt}
\usepackage{authblk}
\usepackage{graphicx}
\usepackage[space]{grffile}
\usepackage{latexsym}
\usepackage{textcomp}
\usepackage{longtable}
\usepackage{tabulary}
\usepackage{booktabs,array,multirow}
\usepackage{amsfonts,amsmath,amssymb}
\providecommand\citet{\cite}
\providecommand\citep{\cite}
\providecommand\citealt{\cite}
% You can conditionalize code for latexml or normal latex using this.
\newif\iflatexml\latexmlfalse
\providecommand{\tightlist}{\setlength{\itemsep}{0pt}\setlength{\parskip}{0pt}}%
\AtBeginDocument{\DeclareGraphicsExtensions{.pdf,.PDF,.eps,.EPS,.png,.PNG,.tif,.TIF,.jpg,.JPG,.jpeg,.JPEG}}
\usepackage[utf8]{inputenc}
\usepackage[ngerman,english]{babel}
\usepackage{float}
\begin{document}
\title{Early-onset of deep sternal wound infection after cardiac surgery is
associated with decreased survival: A propensity weighted analysis}
\author[1,2]{Ramsey Elsayed}%
\author[2]{Joseph N. Carey}%
\author[2]{Robbin Cohen}%
\author[2]{Mark Barr}%
\author[2]{Craig Baker}%
\author[2]{Vaughn Starnes}%
\author[2]{Michael Bowdish}%
\affil[1]{University of Southern California (USC)}%
\affil[2]{University of Southern California Keck School of Medicine}%
\vspace{-1em}
\date{\today}
\begingroup
\let\center\flushleft
\let\endcenter\endflushleft
\maketitle
\endgroup
\selectlanguage{english}
\begin{abstract}
Objectives: To compare outcomes after the development of early ([?]30
days) versus delayed (\textgreater{}30 days) deep sternal wound
infection (DSWI) after cardiac surgery. Methods: Between 2005 and 2016,
64 patients were treated surgically for DSWI following cardiac surgery.
Thirty-three developed early DSWI, while 31 developed late DSWI. Mean
follow up was 34.1 \selectlanguage{ngerman}± 32.3 months. Results: Survival for the entire
cohort at 1, 3, and 5 years was 93.9, 85.1, and 80.8\%, respectively.
DSWI diagnosed early and attempted medical management were strongly
associated with overall mortality (hazard ratio (HR), 25.0 and 9.9; 95\%
confidence intervals (CI), 1.18-528 and 1.28-76.5; p-value 0.04 and
0.04, respectively). Survival was 88.1, 77.0, 70.6 and 100, 94.0 and
94.0\% at 1,3, and 5 years in the early and late DSWI groups,
respectively (Log-rank = 0.074). Those diagnosed early were more likely
to have a positive wound culture (odds ratio (OR), 0.06, 95\% CI
0.01-0.69, p=0.024) and diagnosed late were more likely to be female (OR
8.75, 95\% CI 2.0-38.4, p=0.004) and require an urgent DSWI procedure
(OR 9.25, 95\% CI 1.86-45.9, p=0.007). Both early diagnosis of DSWI and
initial attempted medial management were strongly associated with
mortality (hazard ratio 7.48, 95\% CI 1.38-40.4, p=0.019 and hazard
ratio 7.76, 95\% CI 1.67-35.9, p=0.009, respectively). Conclusions:
Early aggressive surgical therapy for deep sternal wound infection after
cardiac surgery results in excellent outcomes. Those diagnosed with DSWI
early and have failed initial medical management have increased
mortality.%
\end{abstract}\selectlanguage{ngerman}%
\sloppy
\textbf{Introduction}
The historical incidence of deep sternal wound infection (DSWI) after
sternotomy for cardiac surgery ranges between 0.2 and 8\%, with most
recent estimates demonstrating a persistent risk despite implementation
of quality improvement initiatives of 1-2\%{[}1-4{]}. Although
relatively uncommon, this complication is associated with increased
duration of hospital stay or readmission, mortality, significant
morbidity and costs{[}5,6{]}.
Management options for deep sternal wound infection are typically
individualized based on the extent of infection, offending organisms,
and clinical status of the patient. Treatment can include combinations
of antibiotic therapy, wound debridement, negative pressure wound
therapy, partial or total sternal resection, and soft tissue flap
reconstruction{[}7-11{]}. The optimal treatment for these infections
remains controversial and highly individualized based on surgeon and
treatment center{[}7{]}. Although a multitude of risk factors for deep
sternal wound infection have been identified, the timing of onset of
infection as a risk factor for mortality is not well characterized. The
present study examined outcomes for deep sternal wound infection after
cardiac surgery based on the time of onset.
\textbf{Materials and Methods}
\emph{Patient Selection, data collection, and endpoints}
This was a systematic retrospective cohort study of consecutive patients
undergoing surgical management of deep sternal wound infection following
cardiac surgery between January 2006 and December 2016. The
Institutional Review Board of the University of Southern California
Health Sciences Campus approved this study (HS-17-00053) and waived the
requirement for patient consent.
Patients were treated for DSWI at a single center (Keck Hospital of
USC), however, patients transferred from other institutions were
included. All patients were required to have undergone cardiac surgery
via a full sternotomy. Partial sternotomies, other less invasive
approaches, and sternotomies for non-cardiac surgery were not included.
No time limit was placed on the interval between previous sternotomy and
diagnosis of deep sternal wound infection. Patients with sterile sternal
wound dehiscence who only required sternal rewiring were excluded. The
follow-up period closed April 30\textsuperscript{th}, 2017.
Deep sternal wound infection was defined according to the mediastinitis
category as reported in the Centers for Disease Control and Prevention
Guidelines{[}12{]}. The diagnosis of deep sternal wound infection
required one of the following: positive cultures from the mediastinal
area, evidence of infection during surgical exploration, or one of the
following signs or symptoms with no other recognized cause: fever, chest
pain or sternal instability, and either purulent drainage from the
mediastinal area, positive cultures in blood or the mediastinal area, or
mediastinal widening on chest x-ray{[}12{]}.
The cohort was divided into two groups based on the timing of onset of
deep sternal wound infection after the index cardiac procedure. An early
infection was defined as those occurring less than 30 days from the
index procedure, while a late infection was defined as those occurring
greater than or equal to 30 days from the index procedure.
Patient baseline demographics, operative characteristics for the index
cardiac procedure and all procedures related to the DSWI, and
perioperative outcomes were identified through the USC Cardiothoracic
Surgery Database and The Society of Thoracic Surgeons (STS) Adult
Cardiac Surgery Database. All medical records from our electronic
medical record were reviewed. Mortality was confirmed through clinical
follow-up, direct patient (or family) or direct provider contact.
Follow-up was 100\% complete. The primary endpoint was mortality.
\emph{Management Approach}
Once a diagnosis of DSWI is suspected, broad spectrum antibiotics were
started and a CT of the chest obtained.~Fluid resuscitation, nutritional
evaluation, and cardiac~optimization~were instituted prior
to~aggressive~and prompt surgical debridement. Initial exploration,
drainage and irrigation of the mediastinum were~followed~by radical
debridement of all devitalized tissue and removal of sternal wires and
plates. Cultures were routinely obtained. Sternectomy was aggressively
performed for severe infections. Depending on the extent of the
infection and the patients clinical condition,~staged debridement~with
open chest or negative pressure suction was considered versus immediate
flap coverage. Those with overt sepsis, clinical instability, or
extensive infection were typically treated with staged negative pressure
wound therapy.~Aggressive nutritional replacement was instituted, with
enteral feeding if necessary. Once stabilized,~and the debridement
completed, soft tissue coverage was performed~by our plastic surgeons.
The extent of flap coverage was dependent on the extent of the
debridement and included pectoralis myocutaneous advancement, pectoralis
rotation, and omental flaps. Omental flaps were often selected in the
case of mediastinal grafts or extensive dead space. Skin grafts and free
flaps were considered in~patients~with inadequate skin coverage.
\emph{Statistical Analysis}
Data analyses were performed using SAS 9.4 software (SAS institute,
Cary, NC). Figures of odds and hazard ratios were prepared using Prism 7
(GraphPad Software, Inc., La Jolla, CA, USA). Data are presented for the
overall cohort, and the two groups stratified by timing of onset of deep
sternal wound infection after index cardiac procedure. Categorical data
were summarized as the number and percentage of patients. For continuous
data, the mean value and standard deviations were reported for normally
distributed data while the median value and interquartile ranges were
reported for non-normally distributed data. Continuous variables were
compared between the two groups using either Student's t-test or the
Wilcoxon rank sum test, depending on data normality. Categorical
variables were compared using the chi square or Fisher's exact test,
depending on expected values. Unadjusted survival was assessed by
Kaplan-Meier methods for the overall cohort and various strata. Between
strata comparisons were made by log-rank. Multivariable Cox proportional
hazard modeling was used to model predictors of death in the overall
cohort.
Multivariable logistic regression was performed to determine which
preoperative variables were predictive of onset of DSWI within 30 days
of index cardiac procedure. Early onset DSWI was modeled as the
dependent variable and the following independent variables were included
in the model based on univariate analysis, the medical literature, and
biologic plausibility: sex, smoking (yes vs. no), previous cardiac
surgery to the index operation for the DSWI (yes vs. no), positive wound
culture at time of initial DSWI operation (yes vs. no), and urgency of
initial DSWI operation (yes vs. no). The C-statistic and Hosmer-Lemeshow
goodness-of-fit test were reported for the model.
To estimate the impact of timing of onset of deep sternal wound
infection after cardiac surgery on subsequent mortality, the
multivariable regression model was used to calculate a subject's
propensity for having an early onset DSWI. Propensity scores were then
entered into a separate Cox proportional hazard regression model to
calculate adjusted hazard ratios for estimates of mortality.
\textbf{Results}
\emph{Overall cohort}
A total of 64 patients with a deep sternal wound infection after a
previous cardiac procedure were identified during the study period.
Baseline and index operative, and operative characteristics of the DSWI
procedures are shown in \textbf{Tables 1, 2, and 3} . In general, the
overall cohort was predominantly older men (mean 60 \selectlanguage{ngerman}± 12 years) who were
borderline obese (30.2 ± 7.1 kg/m\textsuperscript{2}), hypertensive,
diabetic, and had coronary artery disease. Almost half were transferred
to our center from another institution for management of the DSWI after
a cardiac procedure elsewhere. The most common index cardiac procedure
was a coronary artery bypass grafting, followed by a valve repair or
replacement. DSWI after an aortic procedure or ventricular assist device
were less common in this cohort.
The median time to diagnosis of DSWI was 30 days (Interquartile Range
(IQR) 13.5 -- 67 days). Positive wound cultures were documented in 54
(84\%) with Gram-positive organisms being the most common. Twenty-one
(33\%) patients had an attempt at medical management with antibiotics
alone, however, all eventually underwent surgical therapy as per study
design. A total of 135 operative procedures for treatment of the DSWI
were performed on these 64 patients. Surgical management was debridement
and wound therapy alone in 13 (20\%), while 51 (80\%) underwent flap
coverage as either a primary or secondary procedure.
Mean follow up was 34.1 ± 32.3 months. Overall survival was 93.9, 85.1,
and 80.8\% at 1, 3, and 5 years respectively (\textbf{Figure 1} ).
Univariate Cox proportional hazard modelling showed only a history of
hypertension was associated with overall mortality (hazard ratio (HR),
0.21, 95\% confidence intervals, 0.05-0.96, p=0.44, see Supplemental
Table 1). A multivariable model of mortality with those variables with
univariate p valves \textless{}0.2 (infection diagnosed more than 30
days after index operation, male gender, history of hypertension, and
attempted medical management) showed infection diagnosed within 30 day
of index cardiac procedure and attempted initial medical management were
strongly associate with overall mortality (hazard ratios 25.0 and 9.9,
respectively), while the absence of hypertension was protective (hazard
ratio 0.10, \textbf{Table 4)} .
\emph{Comparison of cohorts based on timing of DSWI diagnosis}
As time of diagnosis of the DSWI in relation to the index procedure was
significant for mortality on adjusted multivariable Cox proportional
hazard modeling, we divided the cohort into two groups: those in which
the diagnosis of DSWI was made within 30 days of the index cardiac
procedures (33 patients, early onset), and those in which it was made 30
days or more after the index cardiac procedure (31 patient, late onset).
Baseline and index operative, and operative characteristics of the DSWI
procedures of the two cohorts are shown in \textbf{Tables 1, 2, and 3} .
Those with early diagnosis of DSWI were less likely to be female, more
likely to be smokers, and more likely to have an elevated glycated
hemoglobin. The early DSWI group also had longer lengths of stay after
the index hospital procedure.
The median time to infection was 14 days in the early onset group and 76
days in the late onset group. Fifty-four patients (84\%) had positive
wound cultures, and a positive wound culture was more common in the
early DSWI group. Types of organisms did not differ between groups.
Equal numbers of patients had an initial attempt at medical management,
however all eventually underwent surgical therapy as per study design.
The initial DSWI operation was performed urgently after diagnosis more
commonly in the late diagnosis group, and electively in the early group.
Initial flap coverage was performed in similar numbers of in the early
diagnosis group, while delayed flap coverage was more common in the
delayed infection group.
Mean follow up was slightly longer in the late DSWI group (32.0±28.6
versus 36.1±35.8 months, p=0.03). Overall survival was 88.1, 77.0, 70.6
and 100, 94.0 and 94.0\% at 1,3, and 5 years in the early and late DSWI
groups, respectively (Log-rank = 0.074).
\emph{Propensity analysis and adjusted outcomes by cohorts of timing of
DSWI diagnosis}
Given the significant differences in baseline characteristics between
the early and late diagnosis groups, five baseline variables were used
to create a logistic regression model for selection to an early or late
diagnosis of DSWI. \textbf{Figure 3} shows the results of the logistic
regression model predicting selection to the early or late diagnosis of
DSWI group. Those more likely to be in the early diagnosis group were
those with positive wound cultures (odds ratio 0.06, 95\% CI 0.01-0.69,
p=0.24) and those more likely to be in the late diagnosis group were
females (OR 8.75, 95\% CI 2.0-38.4, p=0.004) and those requiring urgent
DSWI procedures (OR 9.25, 95\% CI 1.86-45.9, p=0.007). The area under
the receiver operator curve (c-statistic) was 0.84 and the
Hosmer-Lemeshow goodness-of-fit test was not statistically significant
(p=0.62), suggesting good model discrimination and limited collinearity
and interactions \textbf{(Supplemental Figure 1)} .
Propensity score adjusted multivariable Cox proportional hazard
regression outcomes for mortality are shown in \textbf{Figure 4} . Early
diagnosis of DSWI and initial attempted medical management were both
strongly associated with increased mortality (hazard ratio 7.48, 95\% CI
1.38-40.4, p=0.019 and hazard ratio 7.76, 95\% CI 1.67-35.9, p=0.009,
respectively). This was independent of initial operation (flap or
negative pressure wound therapy) or whether any flap was eventually
performed.
\textbf{Conclusions}
Deep sternal wound infection (DSWI) after cardiac surgery is known to be
associated with an increased length of stay, readmission and mortality.
Perrault et. al. recently reported outcomes in 5,198 patients enrolled
in a prospective study evaluating infections after cardiac surgery and
their effect on readmissions and mortality for up to 65 days after
cardiac surgery. The cumulative incidence of DSWI (termed mediastinitis
in this report) was 0.79\% and the median time to diagnosis of infection
was 20.6 days. Readmission rates and mortality were five times higher in
the mediastinal infection group.
Admitting that DSWI occurs and has been relatively resistant to quality
improvement initiatives, we sought to examine outcomes after treatment
for DSWI after cardiac surgery. The main findings of our study are that
overall survival with our surgical approach to these infections is
relatively good, and that those diagnosed with DSWI early and those who
have failed initial medical management have increased mortality. As a
tertiary referral center, almost half of the subjects in the study were
transferred to our institution after their index cardiac surgical
procedure and our general management approach is to be surgically
aggressive with these infections given the known increased mortality
risk in this population.
The overall survival in this cohort of patients was 93\% and 81\% at 1
and 5 years, suggesting that our operative approach results in
acceptable outcomes. These findings of survival after treatment of DSWI
are consistent with other reports. Jones et. al. reported a 8.1\% 20
years mortality rate in 409 patients undergoing flap coverage of
DSWI{[}11{]}, while Baillot et. al. reported a 15 years review of 88\%
three years survival of 124 patients undergoing primary negative
pressure wound therapy as treatment for DSWI{[}10{]}. Others have
reported similar results with a variety of treatment
modalities{[}8,13-16{]}.
Risk factors for increased mortality after development of a DSWI in the
overall cohort included early diagnosis of DSWI (within 30 days of index
cardiac procedure) and attempted medical management. The significant
finding of attempted medical management is not entirely surprising given
that by study design we only included those patients who had undergone
surgical procedures for the DSWI. Therefore, we do not know the true
risk of failed medical management of DSWI at our institution as patients
managed successfully with medical therapy alone, are not included in
this analysis.
The role of timing of diagnosis of DSWI was investigated further by
comparing outcomes in those diagnosed early (\textless{}30 days) and
late (\selectlanguage{english}[?]30 days) with DSWI after index cardiac procedure. Those with an
early diagnosis of DSWI were more likely to be male, more likely to be
smokers, and more likely to have elevated glycated hemoglobin levels.
After developing propensity scores from a multivariable logistic model
to predict differences in baseline characteristics between the two
group, male sex, smoking and a positive wound culture were significantly
more common in the early diagnosis group. Propensity adjusted Cox
proportional hazard modeling demonstrated that early diagnosis of DSWI
and an initial attempt at medical management were strongly associated
with mortality (hazard ratio 7.48, 95\% CI 1.38-40.4, p=0.019 and hazard
ratio 7.76, 95\% CI 1.67-35.9, p=0.009, respectively), and that this
effect was independent of the initial operation (flap or negative
pressure wound therapy) or whether any flap was eventually performed.
Early onset infection was more common in male patients, smokers, and
those with a positive wound culture while female sex and the requirement
of an urgent operation was more common in late onset infection. After
adjusting for differences between both groups, those with early onset
deep sternal wound infection had higher mortality, likely reflecting a
greater degree of aggressiveness of these infections.
These findings support our general philosophy that early aggressive
treatment of these infections is optimal. The results suggest that those
with early onset infections would perhaps benefit from early aggressive
surgical management of DSWI. This general approach is also advocated by
others, as Sears et. al. recently demonstrated in a national database
study that delayed flap closure for DSWI is associated with increased
mortality{[}8{]}.
Limitations to our study should be acknowledged. First, the
retrospective nature of the study limits the ability to draw conclusions
regarding causality. Second, our study might be underpowered to draw
definitive conclusions. Lastly, the timing and surgical approach
undertaken are not standardized and are subject to surgeon preference,
reflecting daily clinical practice.
In summary, these results suggest that the early onset of DSWI is
associated with increased mortality and that a high index of suspicion,
early diagnosis, and aggressive treatment of this devastating
complication after cardiac surgery can results in improved outcomes.
\textbf{Figure Legends}
\textbf{Figure 1} . Kaplan-Meier survival analysis of entire cohort with
a deep sternal wound infection with 95\% confidence banks.
\textbf{Figure 2.} Kaplan-Meier survival analysis after deep sternal
wound infection, stratified by time of onset after cardiac surgery,
(\textless{}31 days, solid line) versus late (\textgreater{}30 days,
dashed line).
\textbf{Figure 3.} Forest plot of multivariable logistic regression
model of late (\textgreater{}30 days) or early (\selectlanguage{english}[?] 30 days) onset of deep
sternal wound infection after cardiac surgery. Odds ratio (OR)
\textgreater{} 1.0 favors late infection (bars represent 95\% confidence
intervals).
\textbf{Figure 4.} Forest plot of propensity adjusted multivariable
analysis of mortality. Hazard ratio (HR \textgreater{}1.0 associated
with increased mortality (bars represent 95\% confidence intervals).
Abbreviations: NPWT = negative pressure wound therapy.
\textbf{Table 1. Baseline characteristics of overall cohort and
comparison of early versus late DSWI strata\textsuperscript{1}}\selectlanguage{english}
\begin{longtable}[]{@{}lllllll@{}}
\toprule
& \textbf{Overall (N=64)} & \textbf{Early DSWI (N=33)} & \textbf{Late
DSWI (N=31)} & \textbf{Odds Ratio} & \textbf{95\% CI} &
\emph{p}\textbf{-value}\tabularnewline
\midrule
\endhead
\textbf{Demographics} & & & & & &\tabularnewline
Age (years) & 60.3 \selectlanguage{ngerman}± 12 & 61.6 \selectlanguage{ngerman}± 10.3 & 59 \selectlanguage{ngerman}± 13.6 &
---\textsuperscript{2} & ---\textsuperscript{2} & 0.38\tabularnewline
Female & 23 (36) & 7 (11) & 16 (25) & 3.96 & 1.32-11.8 &
0.01\tabularnewline
Body mass index (kg/m\textsuperscript{2}) & 30.2 \selectlanguage{ngerman}± 7.1 & 30.7 \selectlanguage{ngerman}± 6 & 29.7
\selectlanguage{ngerman}± 8 & ---\textsuperscript{1} & ---\textsuperscript{1} &
0.67\tabularnewline
\textbf{Comorbidities} & & & & & &\tabularnewline
Hypertension & 50 (78) & 25 (76) & 25 (81) & 1.33 & 0.40-4.40 &
0.64\tabularnewline
Diabetes mellitus & 33 (52) & 17 (52) & 16 (52) & 1.00 & 0.38-2.67 &
0.99\tabularnewline
Coronary artery disease & 44 (69) & 23 (70) & 21 (68) & 0.91 & 0.31-2.62
& 0.87\tabularnewline
Congestive heart failure & 12 (19) & 4 (12) & 8 (26) & 2.52 & 0.67-9.43
& 0.16\tabularnewline
Hyperlipidemia & 27 (44) & 13 (41) & 14 (47) & 1.27 & 0.46-3.49 &
0.63\tabularnewline
End-stage renal disease & 14 (22) & 9 (27) & 5 (16) & 0.51 & 0.15-1.74 &
0.28\tabularnewline
Smoking history & 20 (31) & 14 (42) & 6 (19) & 0.32 & 0.10-1.00 &
0.05\tabularnewline
Anemia & 45 (71) & 25 (76) & 20 (67) & 0.64 & 0.21-1.92 &
0.43\tabularnewline
Immunosuppression & 34 (53) & 19 (58) & 15 (48) & 0.69 & 0.25-1.85 &
0.46\tabularnewline
Reoperative sternotomy & 11 (17) & 3 (9) & 8 (26) & 3.47 & 0.82-14.6 &
0.08\tabularnewline
Hemodynamic instability & 16 (25) & 8 (24) & 8 (256) & 1.08 & 0.35-3.37
& 0.89\tabularnewline
Glycated hemoglobin & 6.5 (5.7, 7.4) & 6.9 (6.4, 7.5) & 5.6 (5.2, 6.5) &
---\textsuperscript{2} & ---\textsuperscript{2} & 0.003\tabularnewline
Preoperative ejection fraction & 55 (43, 65) & 55 (50, 65) & 55 (35, 63)
& ---\textsuperscript{2} & ---\textsuperscript{2} & 0.45\tabularnewline
Transfer from outside hospital & 30 (47) & 13 (20) & 17 (27) & 1.86 &
0.69-5.04 & 0.22\tabularnewline
\bottomrule
\end{longtable}
\textsuperscript{1} Values are n (\%), mean (standard deviation), or
median (interquartile range). Odds ratios and 95\% confidence intervals
are also reported for categorical comparisons between strata.
\textbf{Table 2. Index operative procedure and outcomes for overall
cohort and comparison of early versus late DSWI strata}\selectlanguage{english}
\begin{longtable}[]{@{}lllllll@{}}
\toprule
& \textbf{Overall (N=64)} & \textbf{Early DSWI (N=33)} & \textbf{Late
DSWI (N=31)} & \textbf{Odds Ratio} & \textbf{95\% CI} &
\emph{p}\textbf{-value}\tabularnewline
\midrule
\endhead
\textbf{Index cardiac surgery procedure} & & & & & &\tabularnewline
Coronary artery bypass grafting & 37 (58) & 22 (67) & 15 (48) & 0.46 &
0.17-1.28 & 0.14\tabularnewline
Valve repair or replacement & 15 (23) & 8 (24) & 7 (23) & 0.91 &
0.28-2.90 & 0.88\tabularnewline
Aortic procedure & 9 (14) & 3 (9) & 6 (19) & 2.40 & 0.54-10.58 &
0.24\tabularnewline
Ventricular assist device & 3 (5) & 0 (0) & 3 (10) &
---\textsuperscript{2} & ---\textsuperscript{2} & 0.07\tabularnewline
\textbf{Outcomes for index procedure} & & & & & &\tabularnewline
Length of stay (days) & 14.5 (8, 26.5) & 19 (12, 32) & 10 (6, 17) &
---\textsuperscript{3} & ---\textsuperscript{3} & 0.003\tabularnewline
Intensive care length of stay (days) & 4 (0, 12) & 4 (1, 15) & 3.5 (0,
11) & ---\textsuperscript{3} & ---\textsuperscript{3} &
0.53\tabularnewline
Time to infection (days) & 30 (13.5, 67) & 14 (10, 19) & 76 (37, 238) &
---\textsuperscript{3} & ---\textsuperscript{3} &
\textless{}0.0001\tabularnewline
\textbf{Long-Term Outcomes} & & & & & &\tabularnewline
Follow-up (months) & 34.1 \selectlanguage{ngerman}± 32.3 & 32.0 \selectlanguage{ngerman}± 28.6 & 36.1 \selectlanguage{ngerman}± 35.8 &
---\textsuperscript{3} & ---\textsuperscript{3} & 0.03\tabularnewline
Mortality & 7 (11) & 6 (18) & 1 (2) & 0.15 & 0.02-1.33 &
0.05\tabularnewline
\bottomrule
\end{longtable}
\textsuperscript{}
\textbf{Table 3. Microbiological and operative characteristics of DSWI
procedures for overall cohort and comparison of early versus late DSWI
strata}\selectlanguage{english}
\begin{longtable}[]{@{}lllllll@{}}
\toprule
\begin{minipage}[b]{0.12\columnwidth}\raggedright\strut
\strut
\end{minipage} & \begin{minipage}[b]{0.12\columnwidth}\raggedright\strut
\textbf{Overall (N=64)}\strut
\end{minipage} & \begin{minipage}[b]{0.12\columnwidth}\raggedright\strut
\textbf{Early DSWI (N=33)}\strut
\end{minipage} & \begin{minipage}[b]{0.12\columnwidth}\raggedright\strut
\textbf{Late DSWI (N=31)}\strut
\end{minipage} & \begin{minipage}[b]{0.12\columnwidth}\raggedright\strut
\textbf{Odds Ratio}\strut
\end{minipage} & \begin{minipage}[b]{0.12\columnwidth}\raggedright\strut
\textbf{95\% CI}\strut
\end{minipage} & \begin{minipage}[b]{0.12\columnwidth}\raggedright\strut
\emph{p}\textbf{-value}\strut
\end{minipage}\tabularnewline
\midrule
\endhead
\begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
\textbf{Wound culture}\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
54 (84)\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
32 (97)\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
27 (71)\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
0.19\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
0.04-0.96\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
0.04\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
Gram Positive\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
37 (69)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
22 (59)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
15 (41)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.47\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.17-1.29\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.14\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
Gram Negative\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
12 (22)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
5 (23)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
7 (77)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
1.63\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.46-5.8\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.44\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
Gram Positive and Negative\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
3 (6)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
2 (67)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
1 (33)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.52\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.04-6.00\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
1.00\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
Other\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
2 (4)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
2 (100)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.94\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.86-1.02\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.49\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
\textbf{Timing of DSWI procedure}\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
Urgent (\textless{} 24 hours of diagnosis of DSWI)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
20 (31)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
6 (18)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
14 (45)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
3.70\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
1.19-11.5\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.03\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
Elective ([?] 24 hours of diagnosis of DSWI)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
44 (69)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
27 (82)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
17 (55)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.27\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.08-0.83\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.02\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
\textbf{Initial Attempt at Medical Management}\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
21 (33\%)\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
10 (30)\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
11(35)\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
1.27\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
0.44-3.60\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
0.66\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
\textbf{Requirement of a soft tissue flap}\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
51 (80)\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
31 (61)\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
20 (39)\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
0.60\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
0.22-1.65\strut
\end{minipage} & \begin{minipage}[t]{0.12\columnwidth}\raggedright\strut
0.33\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
Initial\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
23 (45)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
13 (57)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
10 (43)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.84\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.27-2.58\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.76\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
Delayed\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
28 (55)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
17 (61)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
11 (39)\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.84\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.27-2.58\strut
\end{minipage} & \begin{minipage}[t]{0.14\columnwidth}\raggedright\strut
0.76\strut
\end{minipage}\tabularnewline
\bottomrule
\end{longtable}
\textbf{Table 4. Propensity score adjusted multivariate Cox-proportional
hazard model of preoperative risk factors for mortality in overall
cohort of DSWI patients}\selectlanguage{english}
\begin{longtable}[]{@{}llll@{}}
\toprule
\begin{minipage}[b]{0.22\columnwidth}\raggedright\strut
\strut
\end{minipage} & \begin{minipage}[b]{0.22\columnwidth}\raggedright\strut
\textbf{Hazard Ratio}\strut
\end{minipage} & \begin{minipage}[b]{0.22\columnwidth}\raggedright\strut
\textbf{95\% Confidence Interval}\strut
\end{minipage} & \begin{minipage}[b]{0.22\columnwidth}\raggedright\strut
\emph{p}\textbf{-value}\strut
\end{minipage}\tabularnewline
\midrule
\endhead
\begin{minipage}[t]{0.22\columnwidth}\raggedright\strut
\textbf{Variable}\strut
\end{minipage} & \begin{minipage}[t]{0.22\columnwidth}\raggedright\strut
\strut
\end{minipage} & \begin{minipage}[t]{0.22\columnwidth}\raggedright\strut
\strut
\end{minipage} & \begin{minipage}[t]{0.22\columnwidth}\raggedright\strut
\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
Infection within 30 days of index cardiac procedure\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
25.0\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
1.18-528\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
0.039\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
Gender (male)\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
1.60\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
0.13-19.7\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
0.72\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
Hypertension (no)\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
0.10\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
0.13-0.75\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
0.03\strut
\end{minipage}\tabularnewline
\begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
Attempted initial medical management\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
9.88\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
1.28-76.5\strut
\end{minipage} & \begin{minipage}[t]{0.24\columnwidth}\raggedright\strut
0.03\strut
\end{minipage}\tabularnewline
\bottomrule
\end{longtable}
\textbf{Table Legends}
\textbf{Table 1.} Baseline characteristics of overall cohort and
comparison of early versus late DSWI strata
\textbf{Table 2.} Index operative procedure and outcomes for overall
cohort and comparison of early versus late DSWI strata
\textbf{Table 3.} Microbiological and operative characteristics of DSWI
procedures for overall cohort and comparison of early versus late DSWI
strata
\textbf{Table 4.} Propensity score adjusted multivariate
Cox-proportional hazard model of preoperative risk factors for mortality
in overall cohort of DSWI patients
\textbf{Author Contributions:}
Concept/design: RE, JC, RC, MLB, VS, MEB
Data collection: RE, AH, AA
Data analysis/interpretation: RE, MEB
Drafting article: RE, RC, MB, MEB
Critical revision of article: RE, MEB
Approval of article: RE, RC, MLB, VS, MEB
Statistics: RE, MEB
Funding secured by: None
\textbf{References}
{[}1{]} Eklund AM, Lyytik\selectlanguage{ngerman}äinen O, Klemets P, Huotari K, Anttila V-J,
Werkkala KA, et al. Mediastinitis After More Than 10,000 Cardiac
Surgical Procedures. Ann Thorac Surg 2006;82:1784--9.
doi:10.1016/j.athoracsur.2006.05.097.
{[}2{]} Gummert JF, Barten MJ, Hans C, Kluge M, Doll N, Walther T, et
al. Mediastinitis and cardiac surgery--an updated risk factor analysis
in 10,373 consecutive adult patients. Thorac Cardiovasc Surg
2002;50:87--91. doi:10.1055/s-2002-26691.
{[}3{]} Balachandran S, Lee A, Denehy L, Lin K-Y, Royse A, Royse C, et
al. Risk Factors for Sternal Complications After Cardiac Operations: A
Systematic Review. Ann Thorac Surg 2016;102:2109--17.
doi:10.1016/j.athoracsur.2016.05.047.
{[}4{]} Gelijns AC, Moskowitz AJ, Acker MA, Argenziano M, Geller NL,
Puskas JD, et al. Management practices and major infections after
cardiac surgery. J Am Coll Cardiol 2014;64:372--81.
doi:10.1016/j.jacc.2014.04.052.
{[}5{]} Mazzeffi M, Gammie J, Taylor B, Cardillo S, Haldane-Lutterodt N,
Amoroso A, et al. Healthcare-Associated Infections in Cardiac Surgery
Patients With Prolonged Intensive Care Unit Stay. Ann Thorac Surg
2017;103:1165--70. doi:10.1016/j.athoracsur.2016.12.041.
{[}6{]} Greco G, Shi W, Michler RE, Meltzer DO, Ailawadi G, Hohmann SF,
et al. Costs associated with health~care-associated infections in
cardiac~surgery. J Am Coll Cardiol 2015;65:15--23.
doi:10.1016/j.jacc.2014.09.079.
{[}7{]} Lazar, HL, Vander Salm MD T, MD RE, MD DO, MD SG. Prevention and
management of sternal wound infections. J Thorac Cardiovasc Surg
2016;152:962--72. doi:10.1016/j.jtcvs.2016.01.060.
{[}8{]} Sears ED, Momoh AO, Chung KC, Lu Y-T, Zhong L, Waljee JF. A
National Study of the Impact of Delayed Flap Timing for Treatment of
Patients with Deep Sternal Wound Infection. Plastic and Reconstructive
Surgery 2017;140:390--400. doi:10.1097/PRS.0000000000003514.
{[}9{]} Abu-Omar Y, Kocher GJ, Bosco P, Barbero C, Waller D,
Gudbjartsson T, et al. European Association for Cardio-Thoracic Surgery
expert consensus statement on the prevention and management of
mediastinitis. Eur J Cardiothorac Surg 2017;51:10--29.
doi:10.1093/ejcts/ezw326.
{[}10{]} Baillot R, Cloutier D, Montalin L, Côté L, Lellouche F, Houde
C, et al. Impact of deep sternal wound infection management with
vacuum-assisted closure therapy followed by sternal osteosynthesis: a
15-year review of 23499 sternotomies. Eur J Cardiothorac Surg
2010;37:880--7. doi:10.1016/j.ejcts.2009.09.023.
{[}11{]} Jones G, Jurkiewicz MJ, Bostwick J, Wood R, Bried J, Culbertson
J, et al. Management of the Infected Median Sternotomy Wounf with Muscle
Flaps. Annals of Surgery 1997;225:766--78.
{[}12{]} Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition
of health care-associated infection and criteria for specific types of
infections in the acute care setting. Am J Infect Control
2008;36:309--32. doi:10.1016/j.ajic.2008.03.002.
{[}13{]} Raja SG, Berg GA. Should vacuum-assisted closure therapy be
routinely used for management of deep sternal wound infection after
cardiac surgery? Interact Cardiovasc Thorac Surg 2007;6:523--7.
doi:10.1510/icvts.2007.157370.
{[}14{]} Fuchs U, Zittermann A, Stuettgen B, Groening A, Minami K,
Koerfer R. Clinical Outcome of Patients With Deep Sternal Wound
Infection Managed by Vacuum-Assisted Closure Compared to Conventional
Therapy With Open Packing: A Retrospective Analysis. Ann Thorac Surg
2005;79:526--31. doi:10.1016/j.athoracsur.2004.08.032.
{[}15{]} Sjögren J, Gustafsson R, Nilsson J, Malmsjö M, Ingemansson R.
Clinical Outcome After Poststernotomy Mediastinitis: Vacuum-Assisted
Closure Versus Conventional Treatment. Ann Thorac Surg 2005;79:2049--55.
doi:10.1016/j.athoracsur.2004.12.048.
{[}16{]} Petzina R, Hoffmann J, Navasardyan A, Malmsjö M, Stamm C,
Unbehaun A, et al. Negative pressure wound therapy for post-sternotomy
mediastinitis reduces mortality rate and sternal re-infection rate
compared to conventional treatment. Eur J Cardiothorac Surg
2010;38:110--3. doi:10.1016/j.ejcts.2010.01.028.\selectlanguage{english}
\begin{figure}[H]
\begin{center}
\includegraphics[width=0.70\columnwidth]{figures/Fig-1---Overall-Survival-Curve/Fig-1---Overall-Survival-Curve}
\end{center}
\end{figure}\selectlanguage{english}
\begin{figure}[H]
\begin{center}
\includegraphics[width=0.70\columnwidth]{figures/Fig-2---Survival-stratified-by-time-of-onset/Fig-2---Survival-stratified-by-time-of-onset}
\end{center}
\end{figure}\selectlanguage{english}
\begin{figure}[H]
\begin{center}
\includegraphics[width=0.70\columnwidth]{figures/Fig-3---Revised/Fig-3---Revised}
\end{center}
\end{figure}\selectlanguage{english}
\begin{figure}[H]
\begin{center}
\includegraphics[width=0.70\columnwidth]{figures/Fig-4---PS-Cox/Fig-4---PS-Cox}
\end{center}
\end{figure}
\selectlanguage{english}
\FloatBarrier
\end{document}